Report to the FBI Internet Crime Complaint Center (Stable Diffusion content on Signal)

Why this matters for Stable Diffusion content on Signal

• Person specific LoRA adapters trained on a victim's photographs enable repeated NCII generation. LoRA file removal from hosting platforms is a critical and often overlooked lever.
• Signal's zero metadata model is privacy protective for legitimate users but provides no platform side enforcement option for NCII victims. Content can be forwarded to other Signal users without trace.
• Stable Diffusion content distributed through Signal typically reaches between four and twelve additional platforms within 72 hours through reposts, mirrors, and cross platform redistribution.
• Source attribution to Stable Diffusion is critical for report to FBI because it establishes the AI generated nature of the content, which strengthens statutory claims under 47 USC 223a regardless of whether the depicted person ever existed in unaltered form.

How AI generator removal works

What to do, step by step

1. 01
   Preserve evidence before any other action Capture full page screenshots with the URL bar visible, save the highest resolution copy of the content available, and document the Stable Diffusion fingerprint indicators including stable diffusion outputs across versions have known visual fingerprints including characteristic noise patterns in flat regions and recognizable hand and finger artifacts. Full evidence package from the platform takedown including URLs and screenshots
2. 02
   Run a biometric scan to find every copy across platforms Stable Diffusion content distributed on Signal rarely stays there. ScanErase indexes 2.4 billion face embeddings across 200 plus platforms and identifies every current hosting location in a single scan, including additional Stable Diffusion variations.
3. 03
   File the IC3 complaint with full evidence File at ic3.gov with the complete evidence package. The complaint should specifically cite the Stable Diffusion attribution and the Signal distribution. Reference the United Kingdom (Stability AI) jurisdiction of the source service to support investigation routing.
4. 04
   Document the Stable Diffusion source attribution Stable Diffusion outputs across versions have known visual fingerprints including characteristic noise patterns in flat regions and recognizable hand and finger artifacts. This attribution evidence is critical for the legal record and supports stronger statutory claims regardless of which specific action you pursue.
5. 05
   Coordinate with Signal platform removal Signal does not have content moderation tooling because of its end to end encryption and zero metadata architecture. Removal of content requires action against the sender directly.
6. 06
   File the complete evidence record for the legal track 47 USC 223a as the underlying federal NCII statute provides the foundational legal basis for this action. Document every notice sent, every platform response, and every confirmation in a single evidence file. ScanErase produces this Verified Removal Package automatically as your case progresses.

Legal context

• 47 USC 223a as the underlying federal NCII statute
• 18 USC 875 for interstate threat communications when sextortion is involved
• Signal cannot be compelled to remove content because it does not have access to it. The TAKE IT DOWN Act enforcement against Signal is therefore ineffective. The remedy is action against the sender under state NCII statutes.
• Stable Diffusion attribution: The base model is open source and beyond removal action. Action proceeds against the specific fine tune model, the LoRA adapter that enables targeted generation, and the platform that hosts and distributes the harmful adapter.